Analgesic Use, Agranulocytosis, and Aplastic Anemia

To the Editor.— In the Oct 3,1986, issue ofJAMA, some results of an international study on agranulocytosis and aplastic anemia are reported.¹Our comments on this study are confined to agranulocytosis following dipyrone use in the regions of Ulm, West Germany, Berlin, and Barcelona, Spain. The main result of the study was that the risk of agranulocytosis is 23.7 times higher for dipyrone users than for non-users (the rate ratio) and that 27% of all cases of agranulocytosis are caused by dipyrone (the etiologic fraction). These two results are straightforward and replicatable, although the controls are not an unbiased sample of the population, because some controls (eg, those with hernia) were not really healthy in the week before the index day. However, we believe that the authors' most important result, namely, that "the estimated excess risk for any exposure in a one-week period was 1.1 per million," is